checkpoint-1500/trainer_state.json

{
  "best_metric": 0.9162159562110901,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l57-l/checkpoint-1500",
  "epoch": 0.16886187098953057,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.056287290329843524,
      "grad_norm": 150.1642303466797,
      "learning_rate": 4.929640887087696e-07,
      "loss": 0.3486,
      "step": 500
    },
    {
      "epoch": 0.056287290329843524,
      "eval_loss": 1.1265727281570435,
      "eval_runtime": 123.7806,
      "eval_samples_per_second": 15.948,
      "eval_steps_per_second": 1.995,
      "step": 500
    },
    {
      "epoch": 0.11257458065968705,
      "grad_norm": 366.6913146972656,
      "learning_rate": 4.859281774175391e-07,
      "loss": 0.2733,
      "step": 1000
    },
    {
      "epoch": 0.11257458065968705,
      "eval_loss": 0.9742079377174377,
      "eval_runtime": 125.8047,
      "eval_samples_per_second": 15.691,
      "eval_steps_per_second": 1.963,
      "step": 1000
    },
    {
      "epoch": 0.16886187098953057,
      "grad_norm": 555.5016479492188,
      "learning_rate": 4.788922661263087e-07,
      "loss": 0.1851,
      "step": 1500
    },
    {
      "epoch": 0.16886187098953057,
      "eval_loss": 0.9162159562110901,
      "eval_runtime": 126.7606,
      "eval_samples_per_second": 15.573,
      "eval_steps_per_second": 1.949,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 35532,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 540357043592520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}